ALBERT

All Library Books, journals and Electronic Records Telegrafenberg

X
feed icon rss

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
Filter
  • Springer  (106)
  • 2020-2024
  • 1995-1999  (100)
  • 1930-1934  (6)
  • 1
    Electronic Resource
    Electronic Resource
    On neural network design — Part I: Using the MVQ algorithm (1998)
    Springer
    Circuits, systems and signal processing 17 (1998), S. 195-218 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In this two-part study we present a new design methodology for neural classifiers. The design procedure utilizes a multiclass vector quantization (MVQ) algorithm for information extraction from the training set. The extracted information suffices to specify the hidden layer in a canonical neural network architecture. The extracted information also leads to the specification of neuron inhibition rules and subsequently the design of the hidden layer-to-output map. In Part I of the study we focus attention on the MVQ algorithm and how it is used to extract information from a training set. The extracted information is referred to as thecodebook. The codebook is used to directly specify the hidden layer. This specification can take the form of a perceptron layer, a radial basis layer, or a heterogeneous layer involving a mixture of neuron types. These and otherh-layer specifications are determined directly from the same extracted information. The MVQ codebook also suffices to scale the activation function of each neuron. In Part II we consider the nonsimplistic hidden layer-to-output map design. We note that the MVQ algorithm, as it extracts information, decomposes the design set into disjoint neighborhoods. For each neighborhood we identify subsets of the hidden layer neurons, which are significant sensors for the neighborhood. For each such subset we construct an output map. Inhibition rules are established to ensure that the proper output map is activated. In benchmark simulations the overall design exhibits excellent performance, to the extent that we are hard pressed to identify bounds on performance, if any.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01202853
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 2
    Electronic Resource
    Electronic Resource
    Neural network training enhancement (1996)
    Springer
    Circuits, systems and signal processing 15 (1996), S. 467-480 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In this study we explore the use of nonlinear embedding maps to expand the dimension of the input space. The efficacy of such maps to speed training and to enhance performance is illustrated through several examples. A natural connection to nonlinear synaptic interconnects is also developed.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01183155
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 3
    Electronic Resource
    Electronic Resource
    Function emulation using MVQ neural networks (1999)
    Springer
    Circuits, systems and signal processing 18 (1999), S. 315-329 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In earlier studies a multiclass vector quantization (MVQ)-based neural network design was explored for pattern classification. We reconsider that design here in the context of function emulation. With proper adjustment, the MVQ design demonstrates excellent performance. Moreover, the design algorithms sense discontinuities in the data and replicate them in the network.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01200785
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 4
    Electronic Resource
    Electronic Resource
    The approximating capability of fast forms (1999)
    Springer
    Circuits, systems and signal processing 18 (1999), S. 365-376 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In earlier studies the concept of a fast form was generalized to a format that unified such discrete transform examples as the FFT, the FCT, the FST, and the FHT. In this study we consider the approximation of arbitrary linear maps by fast forms. Using simulation we evaluate the approximation capabilities of the generalized fast form.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01200788
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 5
    Electronic Resource
    Electronic Resource
    Auxilliary computations for perceptron networks (1996)
    Springer
    Circuits, systems and signal processing 15 (1996), S. 51-69 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In this study we consider a multilayer perceptron network with sigmoidal activation and trained via the backpropagation algorithm. The output of all neurons is collected and a simple linear regression is performed. It is shown that untrained networks with randomly chosen coefficients perform comparably with fully trained networks. This result casts a new light on the role of activation functions, the impact of dimensionality, and the efficacy of training algorithms such as backpropagation.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01187693
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 6
    Electronic Resource
    Electronic Resource
    On neural network design part II: Inhibition and the output map (1998)
    Springer
    Circuits, systems and signal processing 17 (1998), S. 613-635 
    Details
    ISSN: 1531-5878
    Source: Springer Online Journal Archives 1860-2000
    Topics: Electrical Engineering, Measurement and Control Technology
    Notes: Abstract In this two part study, we presented a new design methodology for neural classifiers. The design procedure utilizes a multiclass vector quantization, MVQ, algorithm for information extraction from the training set. The extracted information suffices to specify the hidden layer in a canonical neural network architecture. The extracted information also leads to the specification of neuron inhibition rules and subsequently to the design of the hidden layer to output map. In part I of that study, we focused attention on the MVQ algorithm and how it is used to extract information from a training set. The extracted information is used to directly specify the hidden layer. In part II, we consider the non-simplistic hidden layer to output map design. We note that the MVQ algorithm, as it extracts information, decomposes the design set into disjoint neighborhoods. For each neighborhood we identify subsets of the hidden layer neurons which are significant sensors for the neighborhood. For each subset we construct an output map. Inhibition rules are established to assure that the proper output map is activated. In benchmark simulations, the overall design exhibits performance, to the extent that we are hard pressed to identify bounds on performance, if any.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01203108
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 7
    Electronic Resource
    Electronic Resource
    Vegetable oils in fermentation: beneficial effects of low-level supplementation (1998)
    Springer
    Journal of industrial microbiology and biotechnology 21 (1998), S. 203-207 
    Details
    ISSN: 1476-5535
    Keywords: Keywords: fermentation; vegetable oils; tetracycline; Streptomyces aureofaciens
    Source: Springer Online Journal Archives 1860-2000
    Topics: Biology , Process Engineering, Biotechnology, Nutrition Technology
    Notes: To evaluate their potential to enhance fermentation performance, vegetable oils were investigated in a model tetracycline fermentation. With sucrose as the carbon source, the fermentation efficiency of Streptomyces aureofaciens (ATCC 10762) was enhanced by the inclusion in the medium of low levels of vegetable oil. Soybean and sunflower oils significantly improved the rate of sucrose consumption and tetracycline production suggesting that oil is an excellent adjuvant for improving fermentation productivity. For optimum benefit, the dosage level was critical. Little difference was observed between crude and refined oils. These data contribute to the assessment of industrially available fermentation feedstocks, and to the development of new feedstock products for specific fermentation applications.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1038/sj.jim.2900571
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 8
    Electronic Resource
    Electronic Resource
    The past and future of the society for industrial microbiology, (Volume 1) (1999)
    Springer
    Journal of industrial microbiology and biotechnology 22 (1999), S. 199-205 
    Details
    ISSN: 1476-5535
    Source: Springer Online Journal Archives 1860-2000
    Topics: Biology , Process Engineering, Biotechnology, Nutrition Technology
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1038/sj.jim.2900632
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 9
    Electronic Resource
    Electronic Resource
    Intermolecular interaction and conformation in poly(ether ether ketone)/poly(ether imide) blends — An infrared spectroscopic investigation (1996)
    Springer
    Journal of polymer research 3 (1996), S. 151-158 
    Details
    ISSN: 1572-8935
    Keywords: Blends ; PEEK ; PEI ; Infrared spectroscopy ; Conformation ; Intermolecular interaction
    Source: Springer Online Journal Archives 1860-2000
    Topics: Chemistry and Pharmacology , Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics , Physics
    Notes: Abstract The intermolecular interaction and the conformation in miscible blends of poly(ether ether ketone) (PEEK) and poly(ether imide) (PEI) have been investigated by Fourier-transform infrared (FTIR) spectroscopy. The intensity of the C=O out-of-phase stretching (1725 cm−1) of PEI shows a minimum at 70 wt% PEI, whereas that of the C=O in-phase stretching (1778 cm−1) is not perturbed by blending. These intensity variations have been attributed to the effect of blending on the coplanarity of the two imide rings bridged by the phenylene group. Change in coplanarity of these two imide rings alters the intensity of the C=O out-of-phase stretching, but it can not affect the intensity of the C=O in-phase stretching. When the two imide rings are perpendicular to each other, the intensity of the C=O out-of-phase stretching is shown to reach the minimum, corresponding to the observation at 70 wt% PEI. The difference spectra (blend - PEEK - PEI) reveal that the bands associated with the diphenyl ether groups in PEEK are modified by blending with PEI. It is proposed that the favorable interaction takes place between the oxygen lone-pair electrons of the ether group in PEEK and the electron-deficient imide rings in PEI.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF01494524
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 10
    Electronic Resource
    Electronic Resource
    Spherulitic growth kinetics in miscible blends of poly(ether ether ketone) and poly(ether imide) (1999)
    Springer
    Journal of polymer research 6 (1999), S. 21-26 
    Details
    ISSN: 1572-8935
    Keywords: Poly(ether ether ketone) ; Poly(ether imide) ; Blend ; Spherulite growth rate ; Surface free energy
    Source: Springer Online Journal Archives 1860-2000
    Topics: Chemistry and Pharmacology , Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics , Physics
    Notes: Abstract Growths of poly(ether ether ketone) (PEEK) spherulites from both pure melt and its miscible blends with poly(ether imide) (PEI) have been studied by polarized optical microscopy. The nucleation density of PEEK spherulites was depressed upon blending with PEI, which can be attributed to the reduction in degree of supercooling arising from equilibrium melting point depression. A modified Lauritzen-Hoffman (L-H) theory was adopted to analyze the growth kinetics. Regime III-II transition was observed with the transition temperature decreasing with increasing PEI composition. Assuming free rotations of the virtual bonds in PEEK molecule, the side surface free energy of 12.0 erg/cm2 was calculated from the characteristic ratio. The fold surface free energy of 188 erg/cm2 and work of chain folding of 12.3 kcal/mol were then obtained from the modified L-H analysis.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/s10965-006-0067-2
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...